In order to solve the situation that the gray wolf algorithm is easy to fall into the local extreme value during the path planning process, this paper proposes an improved gray wolf algorithm incorporating the artificial potential field. Firstly, the convergence factor is improved by a nonlinear strategy to ensure the diversity of populations in the search process; secondly, the gray wolf position update strategy is improved by combining with the beetle antennae search strategy update method to improve the searching ability of the algorithm; finally, the gray wolf algorithm and artificial potential field are fused to improve the searching efficiency of the algorithm as well as the security of the paths. The experimental results show that the improved gray wolf algorithm reduces; the number of robot steering while ensuring a shorter path length.